<![CDATA[West Bandung Regency, West Java Province, is one of the areas with a significant earthquake risk due to the activity of the Lembang Fault and regional geological conditions. Earthquakes that occur suddenly may cause serious impacts on human safety, infrastructure, and activities in campus environments and public buildings. The limited availability of local, real-time, and easily accessible earthquake early warning systems highlights the need for an effective mitigation technology based on the Internet of Things (IoT). This study aims to design and implement an ESP32-based earthquake early warning system prototype integrated with Telegram notifications. The proposed system employs an MPU6050 sensor to measure vibration acceleration along the X, Y, and Z axes, while an SW-420 sensor is used as an initial vibration detector. The acceleration data are processed using the Root Mean Square (RMS) method and the Short Time Average/Long Time Average (STA/LTA) ratio to classify vibration conditions into four warning levels: safe, alert, warning, and danger. The classification results are displayed through local indicators and transmitted in real time to Node-RED using the MQTT (Message Queuing Telemetry Transport) protocol, then forwarded to users through Telegram notifications. The experimental results show that the system is capable of detecting vibration with a success rate of approximately 95%, a classification accuracy of 92%, and a data transmission delay of less than one second. Therefore, the proposed system has the potential to be implemented as a low-cost, responsive, and easily deployable earthquake early warning solution in earthquake-prone areas.]]>
Copyrights © 2026
